Question

我在Ubuntu系统下运行带有CPU版本的Tensorflow 1.9和Python 3.6的python object_detection / builders / model_builder_test.py，存在NameError：未定义名称“ xrange”，有人知道为什么会这样以及如何处理它？谢谢。这是我遵循的指南。

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md

Traceback (most recent call last):
  File "object_detection/builders/model_builder_test.py", line 311, in test_create_ssd_resnet_v1_fpn_model_from_config
    model = model_builder.build(model_proto, is_training=True)
  File "/home/tensorflow/models/research/object_detection/builders/model_builder.py", line 110, in build
    add_background_class)
  File "/home/tensorflow/models/research/object_detection/builders/model_builder.py", line 214, in _build_ssd_model
    ssd_config.anchor_generator)
  File "/home/tensorflow/models/research/object_detection/builders/anchor_generator_builder.py", line 91, in build
    cfg.normalize_coordinates
  File "/home/tensorflow/models/research/object_detection/anchor_generators/multiscale_grid_anchor_generator.py", line 61, in __init__
    for scale in xrange(scales_per_octave)]
NameError: name 'xrange' is not defined

Answer 1

xrange()已从python3中删除，并由range()取代。改为使用range()（其工作原理与xrange()完全相同）。

但是，如果需要在range()中使用python2函数（本例中不需要），请使用xrange()，因为range()返回了一个列表发电机（如@ xdurch0在评论中所述）。

NameError：运行Tensorflow对象检测API时未定义名称'xrange'

1 个答案: